A meta-analysis finds that 3- to 5-minute intervals produce the biggest VO2max boost.
Every once in a while, I get a reminder that Twitter really can be useful. In this case, it was a tweet from Andy Jones pointing out an interesting detail in a new study that I might otherwise have missed. The study is freely available in the open-access journal PLoS ONE, from Michael Joyner of the Mayo Clinic and his colleagues; it’s a meta-analysis of 37 studies in which high-intensity interval training was used to increase VO2max.
The main and unsurprising finding is that intense intervals increase VO2max, and likely do so more effectively/efficiently than continuous exercise; the average increase in the studies examined was 0.51 liters of oxygen per minute. Here’s the distribution of increases for different studies:
Now, there are a couple of interesting points here. The one that Jones pointed out comes from analysis of the tails of the graph above. The intervals training routines used in the 37 studies had all sorts of different parameters — different durations, different interval lengths, different intensities, and so on. But when they focused only on the nine most effective studies, there were some commonalities: in particular, they mostly used intervals lasting between 3:00 and 5:00, longer than the shorter intervals used in some of the less effective studies. So there are some hints, albeit not very rigorous, that „longer“ intervals are most effective. (These top-performing studies also tended to last longer, have higher volumes, and include additional continuous runs too, so we have to be pretty cautious in drawing any specific conclusions.)
The other question that the authors ask is: Would anyone not improve after doing one of these exercise programs? Over the last few years, there has been a lot of discussion about the genetics of trainability. A famous study called the Heritage Study showed that 20 weeks of aerobic training produced a significant average increase in aerobic fitness, but with wide individual variation so that some people didn’t seem to improve at all, or even got worse on some fitness measures. But when Joyner and his colleagues looked at the data ranges for the nine best interval studies, they found that (as far as they could tell from the data available) every individual subject improved. Is it possible that interval training is so effective that it works even for the supposed non-responders?
There are always risks and imperfections in aggregating the data from a bunch of very different studies, so I don’t think we can draw any firm conclusions. But it would be very interesting to see a version of the Heritage Study repeated with a vigorous interval training program. I don’t doubt there would still be great individual variability in response — but if everyone was at least getting better rather than worse, that would be an interesting finding.